INDEX

Explanations

punctuations and variations of colons, indicating elements that outline or introduce lists or elaborations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

roz

-0.15

ummies

-0.15

hi

-0.14

.ObjectId

-0.14

 latter

-0.13

åħ¶ä¸Ń

-0.13

uffers

-0.13

erp

-0.13

uckle

-0.13

alone

-0.13

POSITIVE LOGITS

enames

0.15

ignon

0.14

enance

0.14

 Kota

0.14

ergic

0.14

("'"

0.14

 Given

0.14

enda

0.14

emas

0.13

 belli

0.13

Activations Density 0.049%

punctuations and variations of colons, indicating elements that outline or introduce lists or elaborations

No Comments

No Known Activations